On the Spheroidal State. 89 
length attached 4 cm. from one end, is contracted at the other 
end, and attached to a narrow tube bent twice at right angles, 
so that the two tubes are parallel. The narrow tube termi- 
nates in a small funnel. This apparatus, held vertically in a 
clip, is immersed in a beaker of water, so that the surface of the 
water is about a centimetre below the attached end of the spout- 
like limb, the other end of which extends beyond the edge of the 
beaker. By means of the funnel ether is poured into the nar- 
row tube until it rises in the larger tube to the level of the water 
in the beaker, or a millimetre or two above it. The beaker 
stands on an iron plate which can be heated in any convenient 
way until the temperature of the water in which a thermo- 
meter is placed reaches 30°-32° C. If a narrow pippette with a 
very small opening be now immersed to the depth of 1 or 2 cm. 
in the ether in the large tube and quickly withdrawn to a short 
distance, the drops which fall from it being rapidly cooled by 
evaporation, readily assume the spheroidal state, and often con- 
tinue to float on the warm ether for a long time. The evapora- 
tion from the spheroid is so very slight that some time must 
elapse before any alteration in its size can be detected. On the 
other hand, the warm ether on which the spheroid floats evapo- 
rates very quickly, and its heavy vapour flows out through 
the spout at the side mixed with air which is drawn in at the 
mouth of the large tube. This cool air passing over the surface of 
the spheroid enables it to part with the heat which it is constantly 
receiving from the warm ether by radiation, by the passage of 
heat through the Crookes’s layer on which it is supported, and by 
the precipitation of ether vapour. The ether lost by evaporation 
is replaced from time to time by pouring ether in at the funnel. 
By means of this arrangement I have kept a spheroid about 
5 mm. in its longest diameter, floating for more than an hour 
and a half. 
In the progress of this experiment the sustaining mediuin 
becomes ether vapour alone; it may, however, be experi- 
mentally demonstrated that the presence of a gaseous medium 
other than that derived from evaporation produces the same 
effect. For this purpose it is desirable that the substance em- 
ployed should not be very volatile. It is also important that it 
should be specifically light, as the heavier the spheroid the more 
